(Associate Editor: Yuanlin Song).
Variability and consistency in lung inflammatory responses to particles with a geogenic origin
Version of Record online: 23 DEC 2013
© 2013 The Authors. Respirology © 2013 Asian Pacific Society of Respirology
Volume 19, Issue 1, pages 58–66, January 2014
How to Cite
Zosky, G. R., Boylen, C. E., Wong, R. S., Smirk, M. N., Gutiérrez, L., Woodward, R. C., Siah, W. S., Devine, B., Maley, F. and Cook, A. (2014), Variability and consistency in lung inflammatory responses to particles with a geogenic origin. Respirology, 19: 58–66. doi: 10.1111/resp.12150
- Issue online: 23 DEC 2013
- Version of Record online: 23 DEC 2013
- Accepted manuscript online: 25 JUN 2013 04:00AM EST
- Manuscript Revised: 6 MAY 2013
- Manuscript Accepted: 6 MAY 2013
- Manuscript Revised: 30 APR 2013
- Manuscript Revised: 27 MAR 2013
- Manuscript Revised: 6 MAR 2013
- Manuscript Received: 28 NOV 2012
- particulate matter
Background and objective
Particulate matter <10 μm (PM10) is well recognized as being an important driver of respiratory health; however, the impact of PM10 of geogenic origin on inflammatory responses in the lung is poorly understood. This study aimed to assess the lung inflammatory response to community sampled geogenic PM10.
This was achieved by collecting earth material from two regional communities in Western Australia (Kalgoorlie-Boulder and Newman), extracting the PM10 fraction and exposing mice by intranasal instillation to these particles. The physicochemical characteristics of the particles were assessed and lung inflammatory responses were compared to control particles. The primary outcomes were cellular influx and cytokine production in the lungs of the exposed mice.
The physical and chemical characteristics of the PM10 from Kalgoorlie and Newman differed with the latter having a higher concentration of Fe and a larger median diameter. Control particles (2.5 μm polystyrene) caused a significant influx of inflammatory cells (neutrophils) with little production of proinflammatory cytokines. In contrast, the geogenic particles induced the production of MIP-2, IL-6 and a significant influx of neutrophils. Qualitatively, the response following exposure to particles from Kalgoorlie and Newman were consistent; however, the magnitude of the response was substantially higher in the mice exposed to particles from Newman.
The unique physicochemical characteristics of geogenic particles induced a proinflammatory response in the lung. These data suggest that particle composition should be considered when setting community standards for PM exposure, particularly in areas exposed to high geogenic particulate loads.
mass median aerodynamic diameter
particulate matter <10 μm
Dust particles of geologic origin (geogenic dusts) are widely dispersed in the Earth's atmosphere. In the natural environment, such dusts are derived from unconsolidated sediments and surface soils, and consist mainly of insoluble minerals.
Most geogenic dusts are of medium to coarse grade (approximate range 10–50 μm), with the particles of most interest to human health being those finer than 10 μm (PM10). The airways are vulnerable to injury from any particulate load either from direct irritation or from bioreactive trace elements and compounds that may be adsorbed to or contained within the particles. Mineral dusts have been linked to a range of respiratory conditions, ranging from acute inflammatory reactions to fibrotic changes from more protracted exposure. Other lung disorders linked to or exacerbated by mineral dusts include chronic bronchitis, emphysema and/or small airways disease leading to airflow obstruction.
Inorganic dust inhalation appears to prime lung inflammatory cells in vivo and to enhance their capacity to release toxic oxygen radicals.[6, 7] Minerals such as silica (SiO2) can trigger inflammatory processes, including the generation of reactive oxygen species[8, 9] and cytokines including TNF-α, interleukin (IL)-1β, TGF-β, neutrophil chemoattractants and macrophage inflammatory proteins.[10-13] Importantly, various physical and chemical characteristics of geogenic particles influence the magnitude, duration and characteristics of the inflammatory response.[14, 15] Thus, it is crucial to characterize how such factors may influence the respiratory response, thereby helping to identify which specific particle characteristics may pose the most significant health risk.
Although many occupational health settings have demonstrated that chronic exposure to dusts can cause respiratory diseases,[16, 17] there has been little research on the role of geogenic dusts in a wider community context. A laboratory study using Western Australian iron ore dust showed that such particles can increase interstitial pneumonia and bronchial adenomas in mice and rats. The potential impact of naturally occurring inorganic dust on respiratory health is further highlighted by the presence of particulate matter in lung biopsies from soldiers suffering from exercise limitation due to constrictive bronchiolitis after service in the Middle East.
In order to identify the potential health impacts posed by such exposures, our study assessed the acute inflammatory response in the lung to a single exposure of PM10 particles obtained from two major regional communities in Western Australia; Kalgoorlie-Boulder and Newman. These communities are regularly exposed to geogenic PM10 due to their geology, dry climate, exposure to wind erosion and proximity to open cut mining activities. Using samples of earth materials obtained from these two towns, we examined the effect of particle dose and origin on cellular influx and cytokine production in a controlled laboratory setting using an experimental mouse model.
Eight-week-old female BALB/c mice were obtained from the Animal Resource Centre (Murdoch, Western Australia) and housed in the Telethon Institute for Child Health Research animal house. Mice were exposed to a 12:12 hour light : dark cycle and provided food and water ad libitum.
All experiments were approved by the Telethon Institute for Child Health Research animal ethics committee and strictly conformed to the guidelines of the National Health and Medical Research Council (Australia).
PM10 source and preparation
In order to obtain sufficient quantities of particles to conduct large scale in vivo exposure experiments, we took surface samples as a surrogate for airborne dust (see ‘Chemical composition and validation of the PM10 preparations’ below). For the in vivo exposure study, representative samples were collected from Kalgoorlie (30°44′56′ ′S 121°27′58′ ′E) and Newman (23°21′14′ ′S 119°43′55′ ′E). Kalgoorlie has a population of 28 000 and is located alongside a large open cut gold mine. Newman has an itinerant population of around 6000 and is surrounded by several open cut iron ore extraction sites. In order to obtain a representative sample of earth material, the top 2 cm from a 1 m2 area of surface soil was collected from a number of areas lacking vegetation and open to wind erosion from each town. The samples were transported to the University of Western Australia and the PM10 fraction was extracted using a previously established technique.[20, 21]
Comparison of surface PM10 with airborne PM10
The surface samples were compared to samples collected from dust monitors operated by local authorities. The samples collected by these additional sources were assessed for metal content as per the techniques described below.
Mice (n = 8 in all groups) were exposed intranasally, under light methoxyfluorane anaesthesia, to 10, 30 or 100 μg of PM10 particles from Kalgoorlie and Newman in 50 μL of 0.9 % saline (+ 0.05% Tween-80). Control mice received 50 μL of 0.9 % saline (+ 0.05% Tween-80) alone. Additional groups of mice were exposed to 100 μg of 2.5 μm polystyrene beads (NIST Traceable Polystyrene; Polysciences Inc., Warrington, PA, USA). All exposure preparations were sonicated for 30 min prior to use to prevent particle aggregation.
Bronchoalveolar lavage (BAL)
Mice were euthanized 3, 6, 12, 24 or 168 h after exposure by intraperitoneal injection with an overdose of ketamine/xylazine. Mice were tracheostomized and a polyethylene cannula inserted and secured with suture. A BAL sample was collected by gently instilling and withdrawing 0.5 mL of saline three times, centrifuging the sample, resuspending the cell pellet and staining with Leishman's in order to obtain a differential cell count. The supernatant was stored at −80°C for assessment of cytokines.
Cytokine levels (IL-6, IL-1β, MIP-2, MCP-1) in supernatant samples obtained from the BAL were assessed by enzyme-linked immunosorbent assay according to the manufacturer's instructions (MIP-2, R&D Systems, Minneapolis, MN, USA; IL-6, IL-1β, MCP-1, BD Biosciences, San Jose, CA, USA). These cytokines were chosen to capture the acute inflammatory response (IL-6 and IL-1β) and the expression of chemokines that recruit the major inflammatory cells (macrophages, MCP-1; neutrophils, MIP-2) to the lungs. Total protein content in the supernatant, as a marker of epithelial integrity, was assessed using a colorimetric assay.
The properties of the particles for the exposure protocols were assessed using a number of techniques. Firstly, preparations of all samples (including controls) had endotoxin levels measured using a commercially available limulus amebocyte lysate assay (GenScript, Piscataway, NJ, USA) according to the manufacturer's instructions.
The chemical composition of the stock PM10 samples was measured by inductively coupled plasma-mass spectrometry (Chemistry Centre of Western Australia) for a panel of 12 common metals (Al, As, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, U, Zn) and inductively coupled plasma optical emission spectroscopy (Perkin Elmer Optima 5300DV, Norwalk, CT, USA) for Si. Samples for inductively coupled plasma-mass spectrometry were prepared according to USEPA Method 3051A using mixed acid digestion (HNO3/HCl). Samples for inductively coupled plasma optical emission spectroscopy were fused with X-ray flux and dissolved in HCl. In order to obtain mass weighted estimates of the particle size distributions in the preparations, 2 mL of each sample was aerosolized and drawn through an Andersen Cascade Impacter (Copley Scientific, Nottingham, UK) at 28.3 L/min. The impactor plates were carefully removed and dried at 60°C for 15 min. Pre and post-weights of the impactor plates were obtained using a 5-point balance which allowed the weight of each particle size class to be obtained. The mass median aerodynamic diameter (MMAD) and geometric standard deviation of the particle sizes were calculated using an online tool (http://www.mmadcalculator.com/andersen-impactor-mmad.html).
Non-parametric rank correlation was conducted to validate soil sampling as a proxy for airborne PM10. Inflammatory outcomes were compared between groups by analysis of variance and Holm-Sidak post hoc tests. Due to the scale of the potential number of between group comparisons that were possible, we took a hierarchical approach to the analysis. In the first instance, we compared the saline-exposed group to naïve mice in order to determine if the vehicle had an impact on the outcomes we measured. We then compared the responses in the mice exposed to 10, 30 and 100 μg of geogenic particles with mice exposed to saline or polystyrene particles in order to examine the dose dependent effect/s of the geogenic particles. Finally, we compared the responses between groups of mice exposed to 100 μg of geogenic particles. Data were transformed where necessary to satisfy the assumptions of homoscedasticity and normal distribution of the error terms. Where this was not possible, non-parametric equivalents were used. Data are presented as mean (standard deviation).
Chemical composition and validation of the PM10 preparations
All 13 elements measured by inductively coupled plasma-mass spectrometry/inductively coupled plasma optical emission spectroscopy were detected in the samples used for the in vivo exposure experiments. While the quantity of sample available was not sufficient to determine Si by inductively coupled plasma optical emission spectroscopy in the airborne sample, the surface samples used correlated strongly with the composition of dust obtained from airborne sources for the remaining elements in both cases (Kalgoorlie, rho = 0.75, P = 0.005; Newman, rho = 0.92, P < 0.0001) suggesting that our use of surface PM10 was appropriate for these environments. Si, Al and Fe dominated the samples, although the concentration of these elements differed between sites with Al being the most abundant element in the Kalgoorlie sample after Si, whereas the most abundant element after Si in the Newman sample was Fe. The next most abundant elements were Mn, Cr and Zn, although the order varied by site (Fig. 1).
Endotoxin levels in exposure solutions
The endotoxin levels in the exposure solutions are summarized in Table 1. The lowest endotoxin levels were in the control preparations (e.g. Saline = 2.28 EU/mL) while the highest concentrations were in the 100 μg dust preparations from Kalgoorlie (1272.52 EU/mL) and Newman (8963.12 EU/mL). Using an approximate conversion of 10 EU/ng, our 50 μL preparations had levels of endotoxin ranging from 0.011 ng to 44.8 ng.
|Category||Exposure||Dose||Endotoxin (EU/50_L)||MMAD (μm)||GSD(μm)|
|2.5 μm polystyrene beads||100 μg||0.15|
|Geogenic||Kalgoorlie PM10||10 μg||11.96||1.41||3.62|
|Newman PM10||10 μg||18.16||4.53||1.98|
Physical characteristics and particle size distribution of PM10 preparations
The MMAD was substantially higher in the particles from Newman and the variation in particle size (geometric standard deviation) was narrower than the particles from Kalgoorlie (Fig. 2, Table 1). This was largely due to the dominance of large particles in the Newman sample (Fig. 2).
In vivo responses to the control solution
Exposure to the control solution resulted in a decrease in the number of macrophages 3, 6 and 24 h post-exposure compared to naïve levels (P < 0.05 in all cases) (Fig. 3). There was no difference in the number of neutrophils (Fig. 3) or levels of IL-6, MIP-2, IL-1β, protein between saline-exposed and naïve mice (P > 0.06 in all cases) (Fig. 4).
In vivo responses to the control (polystyrene) particles
The 2.5 μm polystyrene beads caused an increase in the number of macrophages in the BAL 3, 6, 12, 24 and 168 h post-exposure (P < 0.02 in all cases) and an increase in the number of neutrophils 12 (P < 0.001) and 24 h (P < 0.001) post-exposure compared to saline-exposed mice (Fig. 3).
Cytokine and protein responses
Exposure to the polystyrene beads caused a small but detectable increase in MIP-2 in the mice 12 h (P = 0.04) post-exposure (Fig. 4). Mice exposed to 2.5 μm polystyrene beads also had increased protein in the BAL 12 (P < 0.001) and 24 h (P < 0.001) post-exposure. All cytokine and protein levels had returned to naïve levels 168 h post-exposure. There was no discernable response in IL-1β or MCP-1 to the control particle preparations (data not shown).
In vivo responses to the geogenic PM10 preparations
Kalgoorlie – cellular influx
Intranasal exposure to PM10 particles collected from the Kalgoorlie site resulted in a dose dependent influx of cells in the BAL. (Fig. 3). The cellular influx observed at the earlier timepoints in response to the Kalgoorlie PM10 particles was dominated by neutrophils. At all doses, an influx of neutrophils could be detected 6 h post-exposure (10 μg, P = 0.01; 30 μg, P = 0.001; 100 μg, P < 0.001) that persisted until 24 h post-exposure with no difference between exposure groups and the saline-exposed mice 7 days post-exposure (Fig. 3). The magnitude of the response was dose dependent. At 12 h post-exposure, the neutrophil influx in response to 30 (P = 0.001) and 100 μg (P < 0.001) of PM10 was higher than that observed for 10 μg whereas the 100 μg exposure group had higher numbers of neutrophils than the other doses (vs. 10 μg, P = 0.01; vs. 30 μg, P = 0.01) 24 h post-exposure (Fig. 3). In contrast, there was little effect on the number of macrophages in the BAL (Fig. 3). The neutrophil response to 100 μg of Kalgoorlie PM10 exceeded that induced by the polystyrene particles 6 and 12 h post-exposure (Fig. 3).
Kalgoorlie – cytokine and protein responses
Exposure to all concentrations of PM10 from Kalgoorlie resulted in a dose-dependent increase in the level of MIP-2 and IL-6 in the BAL (Fig. 4). The magnitude and duration of the increase in IL-6 was highest in response to 100 μg of PM10 from Kalgoorlie and could be detected at 3 (P < 0.001), 6 (P < 0.001) and 12 h (P = 0.006) post-exposure (Fig. 4). The concentration of IL-6 in the BAL in response to 30 μg of PM10 3 h post-exposure (P = 0.32) was the same as that for 100 μg. IL-6 increased to a lesser extent in the 10 μg group. In the case of MIP-2, the magnitude and duration of the response was the same for the 30 and 100 μg exposure groups, whereas the 10 μg group only had increased levels of MIP-2 6 h post-exposure compared to the saline-exposed mice (P = 0.01) (Fig. 4). There was a significant increase in the protein content of the BAL in the 100 μg exposure group 12 h (P = 0.03) post-exposure and a similar response in the 30 μg group. There was no discernable response in IL-1β or MCP-1 following exposure to PM10 from Kalgoorlie (data not shown).
Newman – cellular influx
Exposure to PM10 particles obtained from the Newman site resulted in a dose-dependent increase in total cell numbers in the BAL (Fig. 3). At the 24 h timepoint, the total cell count was higher in the 100 μg exposure than both the other doses (vs. 10 μg, P < 0.001; vs. 30 μg, P = 0.001) (Fig. 3). The number of macrophages in the BAL of the mice exposed to PM10 from Newman was not different to saline-exposed mice at any timepoint (P > 0.12 in all cases). Thus, the observed increase in total cell counts was primarily attributable to the influx of neutrophils (Fig. 3). The neutrophil response to 100 μg of Newman PM10 exceeded that induced by the polystyrene particles 6, 12 and 24 h post-exposure (Fig. 3).
Newman – cytokine and protein responses
The levels of IL-6 and MIP-2 were increased in the BAL of mice exposed to PM10 from Newman compared to saline-exposed mice (Fig. 4). The magnitude of the increase in IL-6 at the peak of the response was the same in the 30 and 100 μg group (P = 0.20). In contrast, the maximum response was lower in the 10 μg group compared to the other exposure groups (vs. 30 μg, P = 0.005; vs. 100 μg, P < 0.001). At the 3 and 6 h timepoints, the levels of MIP-2 were increased at all three doses compared to saline-exposed mice (P < 0.001 in all cases) (Fig. 4). There was no detectable difference in the magnitude of the response in MIP-2 at these timepoints between PM10 exposure groups. However, the increase in MIP-2 persisted in the 100 μg exposure group and exceeded those of the saline levels 12 h post-exposure (P < 0.001). Protein levels were increased in the mice exposed to 30 μg (P = 0.05) of PM10 12 h post-exposure compared to saline-exposed mice. A similar increase was observed in mice exposed to 100 μg of PM10 (Fig. 4). There was no discernable response in IL-1β or MCP-1 following exposure to PM10 from Kalgoorlie (data not shown).
Comparison of responses to 100 μg particle exposures
One of the aims of this study was to identify particle characteristics that may act to trigger the inflammatory response in the lungs. In order to achieve this, we compared cellular influx and cytokine and protein levels in the BAL of mice exposed to 100 μg of geogenic particles collected from Kalgoorlie and Newman (Fig. S1 available as supporting information online).
The number of macrophages in the BAL of mice exposed to 100 μg of particles from Kalgoorlie was higher than that of mice exposed to particles from Newman 7 days post-exposure although the magnitude of this difference was small (P < 0.001) (Fig. S1 available as supporting information online). The neutrophilia was higher in the mice exposed to particles from Newman compared to both mice exposed to particles from Kalgoorlie (P < 0.001) 24 h post-exposure (Fig. S1 available as supporting information online).
Although levels of MIP-2 were not different between the mice exposed to particles from Kalgoorlie and mice exposed to particles from Newman (P > 0.10 in all cases), mice exposed to particles from Newman had significantly higher levels of IL-6 in the BAL at the 12 h (P < 0.001) time point (Fig. S1 available as supporting information online).
This study has demonstrated that geogenic particles sourced from Kalgoorlie and Newman in Western Australia cause inflammatory responses in the lung. Instillation of particles per se was sufficient to induce an influx of inflammatory cells; however, cytokines associated with detrimental lung effects were significantly higher when the particles were of geogenic origin. Qualitatively, the lung inflammatory response to geogenic particles was similar between sites suggesting that, despite the clear differences in the chemical and physical characteristics of the particles, there was consistency in the inflammatory pathways that were stimulated. Importantly, the magnitude of the response differed considerably highlighting the importance of site specific characterization of the inflammatory properties of environmental particles.
Macrophages are a key component of host defence in the lung to inhaled particles. They are a critical part of the process for expelling foreign particles from the lung and are capable of orchestrating an inflammatory response when necessary. For example, it is well known that macrophages are capable of producing neutrophil chemoattractants leading to the recruitment of neutrophils to the lung. This pattern of response is consistent with our observations in the geogenic exposure groups whereby neutrophilia was associated with increased levels of MIP-2. In line with a proinflammatory response following exposure, both the geogenic particle preparations induced dose-dependent production of IL-6. IL-6, under most conditions, is a proinflammatory cytokine typically associated with acute lung injury.[26, 27] In this respect, the high levels of this cytokine in the Newman and Kalgoorlie groups, even at moderate particle concentrations, demonstrates that these particles may be associated with adverse respiratory outcomes. The lack of production of these cytokines in the control particle exposure groups suggests that it is the unique characteristics of the geogenic particles that were responsible for this observation. However, neutrophilia, similar to that observed in the 2.5 μm polystyrene exposure group may itself be an indicator of an adverse respiratory response. Neutrophils can produce a suite of mediators which, when produced in excess, can cause lung tissue damage. For example the production of neutrophil elastase[28, 29] and reactive oxygen species, both of which are key neutrophil responses, are associated with a wide range of adverse respiratory outcomes. Taken together, these observations suggest that particles in the 2.5 μm size range and those with a mineral origin, such as the geogenic particles, can have a direct impact on lung health even after a single acute exposure.
The physicochemical characteristics were clearly different between the particles collected from Kalgoorlie and Newman. The most striking difference was in the particle size distributions which varied considerably both in terms of the median particle size and the distribution of particles sizes in the sample. It is perhaps surprising then that the characteristics of the inflammatory responses were so consistent between mice exposed to particles from the two sites with a clear influx of inflammatory cells and the production of proinflammatory cytokines. The key difference in the response between groups of mice exposed to these geogenic dust samples was the higher magnitude of the response, particularly the influx of neutrophils and production of IL-6, in the mice exposed to particles from Newman compared to mice exposed to particles from Kalgoorlie. The difference in particle size could have contributed to the increased magnitude of the inflammatory response observed with the Newman particles. Particle size plays a critical role in the respiratory response. For example, there is a size-dependent inflammatory response to silver nanoparticles with smaller particles generating a greater response. In contrast, the size-dependent response to crystalline silica in the fine to ultrafine size range is reversed with larger particles tending to result in a more prolonged effect. The MMAD for the Newman sample was higher than that of the Kalgoorlie sample and the geometric standard deviation was much narrower. As such, and given the primarily geogenic origin of the particles, it is likely that the response we observed is consistent with that for crystalline silica whereby larger particles tend to induce a more prolonged inflammatory response. Indeed, the particle preparations in this instance were dominated (although not exclusively) by silica and the concentration of silica was higher in the Newman particles.
It should be acknowledged, however, that our approach to assessing MMAD and geometric standard deviation was limited. While the use of a cascade impactor remains the gold standard for obtaining mass-weighted estimates of particle size there was a slight disparity between our exposure protocol and the preparation of the particles for the assessment of size. While the mice were exposed by direct instillation of particles in solution the cascade impactor relies on aerosolized particles. As such, the particles themselves would have been contained in aerosolized droplets and as such were likely to have been coated with a film of liquid as they flowed through the impactor. This means that we may have over-estimated the MMAD. However, in this study, we have not attempted to directly correlate MMAD with the response as we only had two sites to compare. Importantly, the relative differences we observed are likely to be reliable such that we are confident that the MMAD was indeed higher in the Newman sample and the variation in particle size was lower compared to the Kalgoorlie sample.
Compared to the Kalgoorlie samples the Newman samples also had a much higher concentration of Fe. Iron is known for its proinflammatory properties when inhaled; in particular in the formation of free hydroxyl radicals leading to oxidative stress. Indeed oxidative stress is associated with neutrophil recruitment and the production of IL-6, both of which were higher in the mice exposed to particles from Newman. It is also possible that the different levels of endotoxin in the preparations may be playing a key role in increased neutrophilia in the mice exposed to particles from Newman. In particular, the level of endotoxin in the 100 μg Newman dust preparations was approximately seven times higher than the equivalent preparation of Kalgoorlie preparations. While the levels of lipopolysaccharide instilled into the mice in the present study were in the submicrogram range; previous studies have been able to detect a small increase in lung neutrophils at doses of lipopolysaccharide around 0.3–0.5 ng/mouse. Clearly, the magnitude of the response in our study was much greater than that observed in other studies specifically designed to examine the effect of lipopolysaccharide. Nonetheless, the higher levels of endotoxin in the Newman samples may have had a synergistic effect on the inflammatory response when combined with the immunogenic properties of the particles themselves.
In comparing the impacts of these particulate exposures, it is important to note that the instillation of the control particles (2.5 μm polystyrene) resulted in moderate inflammatory responses in the lung. Interestingly, the increase in neutrophils associated with the 2.5 μm polystyrene particle exposure group was not associated with an increase in MIP-2. There are several pathways that can lead to neutrophil recruitment to the lung,[35, 36] suggesting that these ‘inert’ particles were activating a different inflammatory pathway in the lung. Thus, exposure of the lung by instillation of particles per se was sufficient to induce an inflammatory response.
In summary, we have shown that particles of geogenic origin sourced from community sites can induce potent inflammatory responses in the lung. The findings imply that accurate monitoring of geogenic material, particularly in communities exposed to high particulate loads, is essential.
- 1Encyclopedia of Environmental Health. Elsevier, Boston, MA, 2011..
- 3Essentials of Medical Geology : Impacts of the Natural Environment on Public Health. Elsevier, Amsterdam, 2005., .
Figure S1 Comparison of inflammatory responses between particles from Kalgoorlie and Newman.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.